Short name:

INARLA

Benchmark:

Description:

n/a

Hardware:

CPU: 8 core, 3.7 GHz, GPU: Tesla K40

Detector:

Public

Processing:

Online

Last submitted:

February 11, 2018 (1 year ago)

Published:

August 09, 2019 at 10:02:30 CET

Submissions:

1

Open source:

No

Project page / code:

n/a

Reference:

H. Wu, Y. Hu, K. Wang, H. Li, L. Nie, H. Cheng. Instance-aware representation learning and association for online multi-person tracking. In Pattern Recognition, 2019.

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
34.770.71.712.5 % 30.0 % 9,85529,1581,1122,848CPU: 8 core, 3.7 GHz, GPU: Tesla K40Public
IDF1ID PrecisionID Recall
42.151.735.4

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
TUD-Crossing73.063.373.30.11353.8 % 0.0 % 122602551
PETS09-S2L240.526.169.91.6427.1 % 11.9 % 7034,657380743
ETH-Jelmoli48.157.173.90.74528.9 % 28.9 % 3299553376
ETH-Linthescher41.348.373.30.419710.2 % 52.8 % 4724,70469164
ETH-Crossing35.248.574.90.3267.7 % 42.3 % 725661222
AVG-TownCentre32.145.767.42.12268.8 % 23.9 % 9553,649250570
ADL-Rundle-120.142.169.26.93218.8 % 9.4 % 3,4633,868100402
ADL-Rundle-336.037.875.02.24415.9 % 18.2 % 1,3835,014109221
KITTI-1632.955.067.41.6170.0 % 5.9 % 32978626113
KITTI-1924.446.966.71.36211.3 % 24.2 % 1,3242,64967326
Venice-136.440.169.01.81729.4 % 11.8 % 8132,05041160

Raw data:

n/a


SNM
INARLA